The authors of the study (read my other comments here) write:

[I]n an average game, one team plays around 15% fewer minutes with black players than their opponent (which roughly corresponds with that team having one fewer black starter). Thus, for this team, the chances of victory under an all-black refereeing crew versus an all-white crew differ by about 0.15*0.226, or around 3.4 percentage points. As such, changing the race of just one referee typically changes the chances of winning by around one percentage point (and the chances of their opponent winning must also change by an offsetting amount).

That is, if a team has one more white starter than its opponent, its chances of winning go up about 1 percent for each white ref. Conversely, given an all-white ref crew, a team’s chances of winning go up 3.4 percent if they have one more white starter.

And:

[I]n our sample, white starters win around 51.8% of their games, but race-norming the refereeing crew [to match the proportion of whites and blacks playing] would likely lower this winning percentage to 50.4%.

Steve Sailer takes this to mean that “NBA teams would win slightly more games if they played whites more.”

That would be absolutely true if you could magically turn some of the black players white. However, as it stands NBA teams simply put out the players they think are best. By hiring and/or putting in more whites, they’d be using players who aren’t the absolute best. Teams have a tradeoff to make between the slight preference from the refs and the drop in quality.

By Robert VerBruggen